Scala code for doing data analysis

SimpleApp.scala

object SimpleApp {
  def main(args: Array[String]) {
    import org.apache.spark.sql.SparkSession
    import org.apache.spark.sql.functions.explode   
    val spark = SparkSession.builder.appName("TripAnalysis").getOrCreate()
    import spark.implicits._
    val results_path = "s3a://results/"
    val paths = "s3a://trips/*"
    val tripDF = spark.read.option("multiline", "true").json(paths)
	// "Explode" the data array into individual rows
    val linksDF = tripDF.select(explode($"data").as("data"))
    val linksDF2 = linksDF.select("data.dbResponse.linkID", "data.absVelocity")
    // create a temporary view using the DataFrame
    linksDF2.createOrReplaceTempView("times")
    /*
      root
      |-- linkID: string (nullable = true)
      |-- absVelocity: double (nullable = true)
    */
    val tDF = spark.sql("SELECT CAST(linkID as LONG), absVelocity from times 
	WHERE linkID IS NOT NULL AND absVelocity IS NOT NULL")
    val groupedDS = tDF.groupBy("linkID")
    val avgsDS = groupedDS.agg(
      "linkID" -> "count",
      "absVelocity" -> "avg"
    ).sort($"linkID".asc)

    avgsDS.coalesce(1).write.
    option("header", "true").
    csv(results_path + "results_49998")
    spark.stop()
  }
}